Summer '21 @ Bloomberg

Downtown Toronto Nightscape

Here I am writing this in early September of 2021, after another 4 months of unpredictable and dynamic changes to our world. At this point, I am not sure if I want to say "let's hope it'll get better in 4 months (or next year)!". Because if I learned anything in the past 1.5 years of craziness is that nobody really knows what is going to happen and it is basically impossible to predict what's next. However, not all is bad, since the last post basically everyone I talk to including myself all are fully vaccinated. Regardless of the delays in vaccine rollout here in Canada compared to the US, we were quick to catch up and now surpassing US's vaccination rate by 10%+. I'm impressed by how fast the medical research and manufacturing community was able to engineer and produce vaccines at this unprecedented speed. Although it is important to note that mRNA technology was not invented recently, it actually have a pretty long history before coming in for the rescue for our vaccines today. Anyways, as per usual, I'll talk a bit about my co-op experience followed by some other random thoughts.

Into the World of Finance @ Bloomberg

Unlike all the other internship roles I had, this time rather than being a software engineer, I was more of a researcher (although my title was still SWE). This means less of making pull requests every day, or even fixing bugs in a repository, but more of reading research papers, trying out different model architectures, and all the good stuff that comes with developing a machine learning model. I did this intentionally to really take a step out of my comfort zone and experience what it's like to do machine learning research. Unsurprisingly, I learned A LOT about how to develop a machine learning model from scratch. Starting with data gathering, formatting, and feature engineering. Which honestly, was the bulk of the work that needed to be done when developing ML models. Having quality data can be argued to be more important than the actual model architecture. And for the first time, I was actually able to use some of the statistics and optimizations knowledge I learned from school into real work! Albeit simple concepts like Pearson correlation, gaussian distribution, and a whole bunch of other statistics stuff. Although I can't talk about the specifics of what I worked on, but I can tell you it was something related to using ML model to build asset pricer. So asset here could mean a currency, stock, bond, etc. Essentially dealing with time series dataset. As with most complicated topics, the more I learn the less I feel like I know about it. This is precisely the Dunning-Kruger Effect at play! Although the problem was certainly challenging, it was really great that I was able to get all the help I needed. From requesting more and more GPU resources, to getting expert opinion on my model architecture.

Okay, now that I talked about my role at Bloomberg let's talk about other aspects of the internship. So as you know, this was unfortunately another remote-only internship. Which was truly a shame, as it was always my goal since HS to be able to work in NYC this summer. But oh well, who knew there would be a pandemic back in HS. What makes it even worse is actually how great the Bloomberg office is!! I was really jealous of interns who was actually able to go in-person and check the office out. Besides that, I really do think the intern coordinators at Bloomberg really tried their best to give us the best experience possible given the remote-only constraint. There was always interesting events and challenges happening. And in fact, my team have actually gotten first place in one of the big challenges, and each of us got a very very very nice suitcase as prize! So that was real fun πŸ˜€. Oh oh, what was also really cool to me was having full access to the Bloomberg terminal. That would otherwise cost about 20,000 USD/year! Needless to say, its something that I probably would never have the chance to use without actually working at Bloomberg. It really is the de facto tool for every financial professional. Just the sheer amount of functionalities, data, and analysis it provides was truly fascinating. And you know what, having all of the cool charts pulled up, really made me feel like I was a wall-street professional 😎. What was also interesting was the other interns that I met. So many of them have such impressive backgrounds, like for example, I had a third of my challenge group teammates being students from MIT. And it was really interesting talking to them. Once again, I regret not being able to meet them in person, as they all were so very nice and smart. Anyways, overall, I think I had a pretty interesting experience at Bloomberg, I learned so much about actual machine learning as well as so much about finance. I highly recommend interning here if you have interests in doing work at the intersection of finance and technology!

State of Our World

Sigh... Don't even get me started. But let's talk about some of the good things first. On the bright side, the majority of Canadians are now fully vaccinated (including myself!). So if you are reading this and you haven't gotten the jab yet, please get it ASAP! Thank you! I'm waiting... still waiting... Okay, got the jab? Great. Now, get ready for some not so bright things. While there's no doubt that vaccination is effective, in fact, it's highly effective. However, it simply isn't sufficient to solve the pandemic alone even if 67% of population (in Canada) are vaccinated. (First of all disclaimer do not have a medical degree, I'm a CS student, all I have to say should not be taken seriously, as I'm not a medical professional in any way shape or form). Given how transmissible the virus is, the more the number of (unvaccinated) people who remain as susceptible hosts to the virus, the more the opportunities for the virus to mutate, which means the need of developing more and more booster vaccine shots. This also mean that, even if 100% of people in developed countries get vaccination, but only 30% of people in rest of the world get vaccinated, this pandemic is not going to go away. So we both need to encourage as many vaccine hesitant people to get vaccinated, as well as ensuring vaccines are distributed relatively evenly across the world. At least we should not have a "vaccine surplus" issue like in the US and Canada while there are parts of the world whom are desperate to get a jab. By sharing our surplus of vaccine supply is not some selfless charity we're doing, but a rather selfish (but completely logical) thing to do. Because only by ensuring everyone is protected against the virus, can we be sure that we are protected. Anyways, this is just my two cents on two seemingly obvious things that we should do. I'm sure that I have missed many details or reasons why that these are not so obvious, so please forgive me πŸ₯Ί.

Watch Me Play Piano!

On a much less serious note, β€Ό shameless self promotion alert β€Ό As a hobby that I developed during the pandemic, I've been recording myself playing some of my favorite piano pieces and uploading them to my Youtube channel 😁. But yeah, given that I have a whole NINE subscribers, I wouldn't call myself a Youtuber, but you know, a boy can dream πŸ˜‚. Here's the latest video so you can get a taste:

Closing Note

Can you believe that it has already been 18 months since the beginning of the pandemic?? That's 18 months of university and in-person internship experience that I'm not going to get back πŸ˜₯. But seriously though, I recognize that while this pandemic has caused me a lot of inconveniences, they are mostly "first world problems" compared to much more terrible and tough problems that others are facing around the world. It really is a luxury to be able to have "not being able to eat out as freely as before" as a "problem". Anyways, I hope everyone can stay safe and act responsibly in ways that are ethical, moral, logical, and caring of others. Until next time, see you around!